bstr: add some utility functions

bstr_strip_ext and bstr_get_ext were taken from find_subfiles.c.

bstr_cut is extended to work like bstr_splice: passing a negative
argument will start counting from the end of the string, e.g.
  bstr_cut("abc", -2) == "bc"

2 files changed, 58 insertions, 0 deletions

diff --git a/bstr.c b/bstr.c
index 0c46b1d9b0..809b02e665 100644
--- a/bstr.c
+++ b/bstr.c
@@ -185,6 +185,19 @@ struct bstr *bstr_splitlines(void *talloc_ctx, struct bstr str)
     return r;
 }
 
+struct bstr bstr_getline(struct bstr str, struct bstr *rest)
+{
+    int pos = bstrchr(str, '\n');
+    if (pos < 0)
+        pos = str.len;
+    if (rest)
+        *rest = bstr_cut(str, pos + 1);
+    str.len = pos;
+    if (str.len > 0 && str.start[str.len - 1] == '\r')
+        str.len -= 1;
+    return str;
+}
+
 void bstr_lower(struct bstr str)
 {
     for (int i = 0; i < str.len; i++)
@@ -233,3 +246,31 @@ int bstr_decode_utf8(struct bstr s, struct bstr *out_next)
         *out_next = s;
     return codepoint;
 }
+
+bool bstr_case_startswith(struct bstr s, struct bstr prefix)
+{
+    struct bstr start = bstr_splice(s, 0, prefix.len);
+    return start.len == prefix.len && bstrcasecmp(start, prefix) == 0;
+}
+
+bool bstr_case_endswith(struct bstr s, struct bstr suffix)
+{
+    struct bstr end = bstr_cut(s, -suffix.len);
+    return end.len == suffix.len && bstrcasecmp(end, suffix) == 0;
+}
+
+struct bstr bstr_strip_ext(struct bstr str)
+{
+    int dotpos = bstrrchr(str, '.');
+    if (dotpos < 0)
+        return str;
+    return (struct bstr){str.start, dotpos};
+}
+
+struct bstr bstr_get_ext(struct bstr s)
+{
+    int dotpos = bstrrchr(s, '.');
+    if (dotpos < 0)
+        return (struct bstr){NULL, 0};
+    return bstr_splice(s, dotpos + 1, s.len);
+}
diff --git a/bstr.h b/bstr.h
index 8b1644cac0..15ddf314f3 100644
--- a/bstr.h
+++ b/bstr.h
@@ -82,8 +82,25 @@ int bstr_decode_utf8(struct bstr str, struct bstr *out_next);
 // On error, -1 is returned. On success, it returns a value in the range [1, 4].
 int bstr_parse_utf8_code_length(unsigned char b);
 
+// Return the text before the next line break, and return it. Change *rest to
+// point to the text following this line break. (rest can be NULL.)
+// Unlike bstr_splitlines, possible \r characters coming from files with CR+LF
+// line breaks are stripped.
+struct bstr bstr_getline(struct bstr str, struct bstr *rest);
+
+bool bstr_case_startswith(struct bstr s, struct bstr prefix);
+bool bstr_case_endswith(struct bstr s, struct bstr suffix);
+struct bstr bstr_strip_ext(struct bstr str);
+struct bstr bstr_get_ext(struct bstr s);
+
+
 static inline struct bstr bstr_cut(struct bstr str, int n)
 {
+    if (n < 0) {
+        n += str.len;
+        if (n < 0)
+            n = 0;
+    }
     if (n > str.len)
         n = str.len;
     return (struct bstr){str.start + n, str.len - n};